Evaluation of primary care physicians' competence in selective skin tumour triage after short versus long dermoscopy training: a randomized non-inferiority trial.

BACKGROUND: Although primary care physicians (PCPs) play a key role in skin cancer screening, their skills in detecting malignant tumours is suboptimal. OBJECTIVES: To determine whether a short dermoscopy e-learning course (4 h) in skin tumour diagnosis for PCPs is non-inferior to a long course (12 h) in selective triage of skin lesions. Secondly, to evaluate whether regular refresher training sessions are necessary to maintain the PCPs' skills in the medium term. METHODS: A randomized 2 × 2 factorial non-inferiority trial was conducted online over an 8-month period among 233 PCPs including 126 certified general practitioners, 94 PCPs in training, and 13 occupational physicians, all without prior advanced dermoscopy training. Participants were randomized 1:1:1:1 to receive short training and mandatory refreshers (n = 58), short training and optional refreshers (n = 59), long training and mandatory refreshers (n = 58), or long training and optional refreshers (n = 58). PCPs' skills were evaluated before training (T0), immediately after training (T1) to test the non-inferiority, and after 5 months (T2) to evaluate the impact of the refreshers. The primary endpoint was the difference in the change of score after short and long training. The non-inferiority margin was set at -28%. RESULTS: Among the 233 randomized participants, 216 (93%) completed T1 and 197 (84.5%) completed T2. For short versus long training, the primary endpoint was 1.392 (95% CI: 0.138; 2.645) in the per-protocol population (p < 0.001) and 1.016 (95% CI: -0.224; 2.256) in the modified intention-to-treat population (p < 0.001). After training, the type of refresher showed no impact on the score (p = 0.840). However, PCPs who completed all refreshers showed the best mean overall score at T2 (p < 0.001). CONCLUSIONS: These findings confirm that short dermoscopy e-learning is non-inferior in training PCPs to triage skin lesions compared to long training. After training, regular refreshers are important to maintain the PCPs' acquired skills over time.

Melanoma Diagnostic Practices of French-Speaking Belgian General Practitioners and the Prospective Study of Their Pigmented Skin Lesion Diagnostic Accuracy and Management.

General practitioners (GPs) are among the main actors involved in early melanoma diagnosis. However, melanoma diagnostic accuracy and management are reported to be insufficient among GPs in Europe. The primary aim of this observational prospective study was to shed light on melanoma diagnostic practices among French-speaking Belgian GPs. The second aim was to specifically analyse these GPs' pigmented skin lesion diagnostic accuracy and management. GPs from the five French-speaking districts of Belgium were asked to complete a questionnaire, before taking part in a melanoma diagnostic training session. First, we assessed the GPs' current melanoma diagnostic practices. Then, their pigmented skin lesion diagnostic accuracy and management were evaluated, through basic theoretical questions and clinical images. These results were subsequently analysed, according to the GPs' sociodemographic characteristics and medical practice type. In total, 89 GPs completed the questionnaire. Almost half of the GPs (43%; CI = [33;54]) were confronted with a suspicious skin lesion as the main reason for consultation once every 3 months, while 33% (CI = [24;43]) were consulted for a suspicious lesion as a secondary reason once a month. Prior to training, one-third of the GPs exhibited suboptimal diagnostic accuracy in at least one of six "life-threatening" clinical cases among two sets of 10 clinical images of pigmented skin lesions, which can lead to inadequate patient management (i.e. incorrect treatment and/or inappropriate reinsurance). This study underlines the need to train GPs in melanoma diagnosis. GPs' pigmented skin lesion diagnostic accuracy and management should be improved to increase early melanoma detection.

Dermatoscopy of combined blue nevi: a multicentre study of the International Dermoscopy Society.

BACKGROUND: Combined blue nevi (CBN) may mimic melanoma and are relatively often biopsied for diagnostic reasons. OBJECTIVE: To better characterize CBN and to compare it with melanoma. METHODS: We collected clinical and dermatoscopic images of 111 histologically confirmed CBN and contrasted their dermatoscopic characteristics with 132 partly blue coloured melanomas. Furthermore, we compared the accuracy of human experts using pattern analysis with a computer algorithm based on deep learning. RESULTS: Combined blue nevi are usually flat or slightly elevated and, in comparison with melanoma, more frequent on the head and neck. Dermatoscopically, they are typified by a blue structureless part in combination with either brown clods (n = 52, 46.8%), lines (n = 28, 25.2%) or skin-coloured or brown structureless areas (n = 31, 27.9%). In contrast with melanoma, the blue part of CBN is more often well defined (18.9% vs. 4.5%, P < 0.001) and more often located in the centre (22.5% vs. 5.3%, P < 0.001). Melanomas are more often chaotic (OR: 28.7, 95% CI: 14.8-55.7, P < 0.001), have at least one melanoma clue (OR: 10.8, 95% CI: 5.2-22.2 P < 0.001) in particular white lines (OR: 37.1, 95% CI: 13.4-102.9, P < 0.001). Using simplified pattern analysis (chaos and clues), two raters reached sensitivities of 93.9% (95% CI: 88.4-97.3%) and 92.4% (95% CI: 86.5-96.3%) at corresponding specificities of 59.5% (95% CI: 49.7-68.7%) and 65.8% (95% CI: 56.2-74.5%). The human accuracy with pattern analysis was on par with a state-of-the-art computer algorithm based on deep learning that achieved an area under the curve of (0.92, 95% CI: 0.87-0.96) and a specificity of 85.3% (95% CI: 76.5-91.7%) at a given sensitivity of 83.6% (95% CI: 72.5-91.5%). CONCLUSION: CBN usually lack melanoma clues, in particular white lines. The accuracy of pattern analysis for combined nevi is acceptable, and histopathologic confirmation may not be necessary in exemplary cases.

Dermatoscopic features of thin (≤2 mm Breslow thickness) vs. thick (>2 mm Breslow thickness) nodular melanoma and predictors of nodular melanoma versus nodular non-melanoma tumours: a multicentric collaborative study by the International Dermoscopy Society.

BACKGROUND: Thin nodular melanoma (NM) often lacks conspicuous melanoma-specific dermatoscopic criteria and escapes clinical detection until it progresses to a thicker and more advanced tumour. OBJECTIVE: To investigate the dermatoscopic morphology of thin (≤2 mm Breslow thickness) vs. thick (>2 mm) NM and to identify dermatoscopic predictors of its differential diagnosis from other nodular tumours. METHODS: Retrospective, morphological case-control study, conducted on behalf of the International Dermoscopy Society. Dermatoscopic images of NM and other nodular tumours from 19 skin cancer centres worldwide were collected and analysed. RESULTS: Overall, 254 tumours were collected (69 NM of Breslow thickness ≤2 mm, 96 NM >2 mm and 89 non-melanoma nodular lesions). Light brown coloration (50.7%) and irregular brown dots/globules (42.0%) were most frequently observed in ≤2 mm NMs. Multivariate analysis revealed that dotted vessels (3.4-fold), white shiny streaks (2.9-fold) and irregular blue structureless area (2.4-fold) were predictors for thinner NM compared to non-melanoma nodular tumours. Overall, irregular blue structureless area (3.4-fold), dotted vessels (4.6-fold) and serpentine vessels (1.9-fold) were predictors of all NM compared to non-melanoma nodular lesions. LIMITATIONS: Absence of a centralized, consensus pathology review and cases selected form tertiary centres maybe not reflecting the broader community. CONCLUSIONS: Our study sheds light into the dermatoscopic morphology of thin NM in comparison to thicker NM and could provide useful clues for its differential diagnosis from other non-melanoma nodular tumours.

Man against machine reloaded: performance of a market-approved convolutional neural network in classifying a broad spectrum of skin lesions in comparison with 96 dermatologists working under less artificial conditions.

BACKGROUND: Convolutional neural networks (CNNs) efficiently differentiate skin lesions by image analysis. Studies comparing a market-approved CNN in a broad range of diagnoses to dermatologists working under less artificial conditions are lacking. MATERIALS AND METHODS: One hundred cases of pigmented/non-pigmented skin cancers and benign lesions were used for a two-level reader study in 96 dermatologists (level I: dermoscopy only; level II: clinical close-up images, dermoscopy, and textual information). Additionally, dermoscopic images were classified by a CNN approved for the European market as a medical device (Moleanalyzer Pro, FotoFinder Systems, Bad Birnbach, Germany). Primary endpoints were the sensitivity and specificity of the CNN's dichotomous classification in comparison with the dermatologists' management decisions. Secondary endpoints included the dermatologists' diagnostic decisions, their performance according to their level of experience, and the CNN's area under the curve (AUC) of receiver operating characteristics (ROC). RESULTS: The CNN revealed a sensitivity, specificity, and ROC AUC with corresponding 95% confidence intervals (CI) of 95.0% (95% CI 83.5% to 98.6%), 76.7% (95% CI 64.6% to 85.6%), and 0.918 (95% CI 0.866-0.970), respectively. In level I, the dermatologists' management decisions showed a mean sensitivity and specificity of 89.0% (95% CI 87.4% to 90.6%) and 80.7% (95% CI 78.8% to 82.6%). With level II information, the sensitivity significantly improved to 94.1% (95% CI 93.1% to 95.1%; P < 0.001), while the specificity remained unchanged at 80.4% (95% CI 78.4% to 82.4%; P = 0.97). When fixing the CNN's specificity at the mean specificity of the dermatologists' management decision in level II (80.4%), the CNN's sensitivity was almost equal to that of human raters, at 95% (95% CI 83.5% to 98.6%) versus 94.1% (95% CI 93.1% to 95.1%); P = 0.1. In contrast, dermatologists were outperformed by the CNN in their level I management decisions and level I and II diagnostic decisions. More experienced dermatologists frequently surpassed the CNN's performance. CONCLUSIONS: Under less artificial conditions and in a broader spectrum of diagnoses, the CNN and most dermatologists performed on the same level. Dermatologists are trained to integrate information from a range of sources rendering comparative studies that are solely based on one single case image inadequate.

Health-related quality of life in patients with melanoma expressed as utilities and disability weights.

BACKGROUND: Few studies about health-related quality of life (HRQoL) in patients with melanoma have expressed their results in terms of utilities or disability weights (DWs). Utilities are required for calculating quality-adjusted life years and therefore for cost-effectiveness analyses. DWs are useful to assess the burden of diseases through disability-adjusted life years. OBJECTIVES: To provide utilities and DWs regarding patients with melanoma. METHODS: The patients were classified into eight groups using four stages based on the 2009 American Joint Committee on Cancer stages, with each stage subdivided into treatment and remission phases. The EuroQoL Five Dimensions Five Levels (EQ-5D-5L) questionnaire was completed by the patients with melanoma to provide a mean utility for each group. In addition to this, the EuroQoL visual analogue scale (VAS) and a validated quality-of-life questionnaire dedicated to patients with melanoma [Functional Assessment of Cancer Therapy Melanoma (FACT-M)] were completed by the same patients in order to compare their results with the obtained utilities. DWs were obtained by calculating, for each patient, the difference between his/her utility and the corresponding sex- and age-specific population norm. RESULTS: A total of 395 questionnaire sets were completed. Utilities and DWs showed significant differences between follow-up groups. Treatment groups had similar utilities and DWs but these results were obtained during different treatment durations and therefore have different weights. The VAS and the FACT-M were found to be less sensitive. Nevertheless, the FACT-M identified some problems not found by the EQ-5D-5L questionnaire. CONCLUSIONS: The EQ-5D-5L questionnaire seems adequate to provide utilities and DWs in patients with melanoma. Lower HRQoL in female patients with melanoma is probably linked to lower HRQoL in the general population.

Availability of digital dermoscopy in daily practice dramatically reduces the number of excised melanocytic lesions: results from an observational study.

BACKGROUND: Digital dermoscopy has been shown to permit an earlier detection of melanoma. However, few studies have investigated its added value in reducing unnecessary excisions in everyday clinical practice. OBJECTIVES: To compare, in daily practice, the efficiency of three dermoscopy methods: dermoscopy alone with little training, dermoscopy alone with adequate training and dermoscopy with adequate training and access to digital dermoscopy, and to confirm the safety of this latter approach. METHODS: Thirty-six dermatologists working without digital dermoscopy were divided into two groups according to their training in dermoscopy. The third group constituted of two dermatologists working in a pigmented lesion clinic with access to the digital dermoscopy technique and eight additional dermatologists working in the same dermatology department. These 46 dermatologists included all presumed melanocytic lesions excised over a period of 1 year. The primary endpoint was the melanoma/nonmelanoma ratio (M/NM-R); secondary endpoints were the ratio of 'problem' naevi to common naevi (PN/CN-R), specificity and sensitivity for the diagnosis of melanoma, in situ/invasive melanoma ratio, and the mean Breslow thickness. RESULTS: In total, 1865 excised lesions, including 231 melanomas, were included. In the digital dermoscopy availability group (DD-G) the M/NM-R was significantly better (1/2.43), as was the PN/CN-R (1/1.48) (P < 0.001 in both cases). The specificity was significantly higher in the DD-G and significantly higher for trained examiners as compared with examiners with little training. More that one-third of all melanomas discovered by digital dermoscopy were in situ, and the mean Breslow thickness was 0.32 mm for the invasive ones. CONCLUSIONS: The reduction of unnecessary excisions when using digital dermoscopy compared with dermoscopy alone in our study suggests that access to digital dermoscopy offers a better management of pigmented lesions in daily practice. The high number of early lesions diagnosed by this technique confirms that its use is safe.

Melanoma screening: results of the first one-day campaign in Belgium ('melanoma Monday').

BACKGROUND: Although the incidence of melanoma is increasing and many informative campaigns have been organized. The general population is still little informed about this tumour. AIMS: To organize a media campaign, with more relevant information and the opportunity for free skin inspections. METHODS: A 'Task Force' organized a media campaign in April 1999 and convinced 65% of the Belgian dermatologists to give up 4 h of their time to do free skin examinations for skin cancer on Monday 26 April 1999; it was called 'Melanoma Monday'. RESULTS: A total 2767 patients were screened. We found 25 melanomas and suspected 59 basal cell carcinomas. In the following 4 weeks another 141 melanomas were found. These 166 melanomas found in one month represent 15-20% of the total number of melanomas per year in Belgium. SUMMARY: A media campaign with relevant information combined with screening opportunities can lead to the early detection of melanomas in a large number of patients and can continue to alert people at risk in the following weeks.

Skin signs in the diagnosis of thallium poisoning.

A 45-year-old man developed a painful and rapidly progressive sensory-motor polyneuropathy associated with confusion and convulsions. This resulted in hypoventilation and led to respiratory failure and coma. A rapid and diffuse alopecia occurred after 3 weeks in the intensive care unit. Examination of hair roots under polarized light detected dystrophic anagen hairs with dark bands caused by empty spaces in the disorganized cortex. These dark zones were originally reported in patients with thallium poisoning and a toxicological investigation confirmed thallium exposure. The classical systemic symptoms and the various dermatological signs are reviewed, and the origins of contamination and physiopathology discussed.